2. UroToday Home
  3. Conference Highlights
  4. AUA 2017
  5. AUA 2017 Prostate Cancer

AUA 2017

AUA 2017: A Systematic Review of Instrumental Variable Analyses Using Instrumental Variable Analyses using Geographic Region as an Instrument in Prostate Cancer Studies

Boston, MA (UroToday.com) The nature of urologic oncologic research does not lend itself well to clinical trials, particularly in the setting of patient-reported outcomes. Given the extensive natural history of prostate cancer, observational studies are increasingly used to establish treatment outcomes, oncological effectiveness, and patient quality of life. Ms. Emily Vertosick of Memorial Sloan Kettering Cancer Center explores the use of “geographic region” as a potential confounder requiring multivariate adjustment in observational studies.

Ms. Vertosick and colleagues conducted a literature review of cancer studies done in the United States and identified 22 eligible studies utilized or adjusted for “geographic region” in multivariate analysis. Interestingly 10 of the 22 studies included in analysis were prostate cancer studies, including those comparing efficacy across the different primary therapies of radical prostatectomy, active surveillance, androgen deprivation therapy, radiotherapy, and minimally invasive surgical technique.

Particularly pertinent to the study was the use of “geographic region” as a surrogate adjustor of “healthcare access”. Treatment rates in high versus low use areas differed from 5% to 31%. While 4/7 covariate models found at least one significant variable associated with treatment prevalence, “geographic region” was generally closer to the null.

By performing a meta-analysis of current publications, treatment variability and outcome inconsistencies were compared across geographic regions. However, despite proposed surrogacy for “healthcare access”, it is advised that geographic region is used with caution. The value of adding geographic region to multivariate analysis may be restricted to a small subset of observational studies and should be further studied across all urologic oncologic studies.

Presented By: Emily Vertosick, MPH, Memorial Sloan Kettering Cancer Center

Authors: Emily Vertosick, Melissa Assel, Andrew Vickers

Written By: Linda Huynh (BS), an assistant research specialist from the University of California, Irvine, on behalf of UroToday.com

at the 2017 AUA Annual Meeting - May 12 - 16, 2017 – Boston, Massachusetts, USA